Talk:Morphological typology

Wiki Education Foundation-supported course assignment

This article is or was the subject of a Wiki Education Foundation-supported course assignment. Further details are available on the course page. Student editor(s): Jpd50616. Peer reviewers: Jpd50616.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 04:28, 17 January 2022 (UTC)

English

It would be nice to see where English is classified in the article, as anyone who is generally unfamiliar with linguistics (as I am) would then have some kind of basis for comparison.

Somewhere between synthetic and analytic, I'd guess, as most modern IE languages(?)... I believe these example languages are chosen for their "purity" of being of a certain type. As an example of how english has different typology for different grammatical functions: "The dog ate the fish" Vs "The fish ate the dog" is purely analytic, whereas strong verb conjugations such as sing,sang,sung is truly synthetic. I believe old english was much more synthetic than modern english, but, like many other IE languages, it evolved into a more analytic language throughout the ages.

Hmmm, maybe the strong verbs are more "fusional" than "synthetic", I must have missed that when I wrote this answer, earlier...

As languages go, English is quite analytic, more so than other Indo-European languages. Inflectional morphology is v. limited, confined to plurals (dog:dogs), the enclitic possessive (dog:dog's), the third person present inflection (John walk-s), the past tense (John walk-ed), the participial and gerunds (walk-ing, walk-ed), the strong verbs (ex, sing, sings, singing, sang, sung), pronouns, and various irregular words. Our derivational morphology, though, is rather complex (and agglutinative rather than fusional) than Anglo-Saxon's, thanks to the influx of words and affixes from the classical languages (ex, anti-, mini-, -ment, ize, all of which are Greco-latin in origin). Most other Indo-European languages are more synthetic, witness German declensions and Spanish conjugations. Spoken French tends towards polysynthesism, even (a recent inovation): Jean, il l'y a acheté la voiture, where il l'y acheté is pronounced as one word (lit. John, he it-there had bought the car, John bought the car there.) In fact, Mandarin Chinese can be analyzed as being more synthetic than English, though both the Hanzi and pinyin obscure it. A majority of words are bisyllabic (pengyou, zuotian, etc) and there's many bound morphemes (all measure words, for example). Since Chinese is so different from Western languages, no plural markings or conjugations, westerners tend to think it's incredibly isolating. English is considered more synthetic than it is because it's Indo-European, a language family tending strongly towards synthesysm. Of course, both Chinese and English are more isolating than synthetic. Generally speaking, an isolating language is one that marks all grammatical relationships purely by syntax, with no morphology at all, whereas a totally synthetic language would mark everything through morphology. Neither extreme exists, of course; it's a continuum. Thai is one extreme, Inuit the other. The terms fusional and agglutinative refer to types of synthesism. A fusional language marks many things in one morpheme. The -a in Rus. doma marks the word as being feminine, singular and genitive (IIRC), whereas an agglutinative language (Turkish, Quechua) would mark each thing separately.

Correction: -a in Russian doma either marks it as masculine, plural, nominative noun (domá) or as masculine, singular, genitive none (dóma). "Dóma" could also be an adverb meaning "at home".

Any other fusional groups?

It seems to me that only Indo-European languages are fusional. Most others seem to be agglutinating or isolating. Are there any other families with fusional tendencies?

Yes. Finno-Ugric and Kartvelian languages tend to be highly fusional. Rmalouf (talk) 01:36, 28 February 2014 (UTC)

Hm, I tend to think of them as largely agglutinative with fusional tendencies. There should be better examples. I was taught these examples as prototypical representations of the traditional types:

• Isolating: Vietnamese (better than Mandarin Chinese)
• Agglutinative: Turkish
• Fusional: Takelma
• Polysynthetic: Greenlandic

I'm aware of the numerous flaws of this simplistic scheme, of course. But in the nineties it was apparently still seen as useful for beginning students of linguistics. --Florian Blaschke (talk) 10:00, 29 May 2015 (UTC)

As I see it, the problem with the category "fusional" is that it conflates three qualities into a general idea of "inseparable morpheme boundaries":

1. Condensing multiple categories into one morpheme, e.g. Slavic noun endings, Romance verb endings
2. Non-concatenative morphology, e.g. Germanic verb and plural ablaut
3. Elision of morpheme boundaries, e.g. Celtic

The thing is, these qualities don't seem to correlate particularly well with one another, and are easy to find examples of outside Indo-European. To name a few, Quechua fuses person/number marking of subject, object, and the past tense; the Semitic languages (Semitic and some Uralic languages are the only commonly cited non-IE fusional languages in the traditional paradigm), Maasai, and Iau inflect primarily via albaut; and Wari' and a number of Na-Dene languages go particularly wild with morpheme elision, and do it very irregularly as well. Yet of course, most of the IE languages don't even feature all of these categories - English, for example, only has ablaut in a subset of its irregular verbs, and neither of the others. So if you're looking for other languages that are in some ways fusional, there are many examples; the disappointing part is that IE, as a group, isn't terribly "fusional" in the first place. Tezero (talk) 16:54, 31 May 2015 (UTC)

Actually, I was taught a definition that posits a 1:1 relationship between morphemes and grammatical categories in agglutinative languages and a one-morpheme-vs.-one-meaning relationship in fusional languages, while in isolating languages the relationship would be one word vs. one meaning. But you are right that reality is much more complicated than that schematic approach, the Quechua example in particular having struck me lately. --Florian Blaschke (talk) 23:03, 31 May 2015 (UTC)

But then, it has long been acknowledged that languages are generally a mixture of these ideal types, and syntheticity/analyticity is best thought of as one of several dimensions on a continuum. English in particular is the classic example of a language which has moved from fusional/synthetic towards highly analytic, almost isolating. The observation that the descendants of many ancient (fairly) synthetic Indo-European languages have moved in the direction of analyticity is what inspired this schema in the first place: it was thought that more synthetic languages were somehow superior and that the move towards analyticity signalled a "decay", and isolating languages such as Mandarin Chinese were particularly primitive, with racist overtones of course – until it was discovered that many technologically primitive cultures in isolated, exotic places have staggeringly and and even dauntingly morphologically complex languages, so much that a common stereotype among laypeople now – at least in the US – seems to be that "simple" languages are superior, as they are often widespread and associated with advanced, urban civilisations, and complicated exotic or ancient languages are mere curiosities unworthy of attention and rightly doomed to extinction; this again has racist overtones and offsets positive attitudes towards Native American languages and cultures in particular. --Florian Blaschke (talk) 23:22, 31 May 2015 (UTC)

Yeah, analyses like that are pretty much doomed to fail because of how typologically diverse the languages of most large regions of the world are. In Europe, for example, you've got English, the North Germanic languages, and possibly Armenian on the analytic end, but Finnish, Georgian, and the Northwest and Northeast Caucasian languages could reasonably be called polysynthetic. Then in the Americas - even though there's a clear tendency toward high synthesis, some languages like the Oto-Manguean and Mura families and Tupi are far on the analytic side of things. Although I don't agree with it to the absolute extent to which it's often posited, the idea that when languages lose complexity in one area, they inevitably gain it in others, explains a lot of how balderdash the scientific racism you bring up is. People's lives are pretty complex, so we need complex frameworks to explain them. Tezero (talk) 19:07, 1 June 2015 (UTC)

Finnish polysynthetic? It doesn't even have polypersonal agreement or noun incorporation, which are characteristic of what is traditionally described as polysynthesis. It's true that you can form some absurdly long word forms in Finnish, but these are never really used even in the most literary Finnish, let alone in spoken Finnish (morphologically complex word forms being considered bookish). It's like those huge compounds you can theoretically form in German (and other Germanic languages, and yes, in Finnish, too). They are intelligible but not really used in practice. Finnish is nowhere like Eskimo languages or Nahuatl in that respect, where these monsters are regularly used. (See Longest words#Finnish.) If you analyse the clitic pronouns as affixes (since word order is fairly rigid), French is a far more obvious candidate for a polysynthetic language in Europe (and so may be other Romance languages), and Egyptian Arabic verbs are effectively polysynthetic too (although they do not seem to have noun incorporation).

Otherwise, I agree. While forager cultures, especially, have little technology and what they have is simple, and their societies are small, they can be really complex socially, with lots and lots of rules and taboos, and also need tons of specialised knowledge about the natural world that most people in technologically advanced, urbanised societies lack. So foraging societies will have countless names for animals and plants and whatnot, which is a real problem for linguists recording their languages since linguists are not usually biologists at the same time. --Florian Blaschke (talk) 06:30, 2 June 2015 (UTC)

Suprasegmental features

Morphemes may also be expressed by internal phonological changes in the root (i.e. morphophonology), such as consonant gradation and vowel gradation, or by suprasegmental features such as stress or tone, which are of course inseparable from the root.

I have trouble understanding this bit. I thought suprasegmentals were not lexically distinctive, and hence separable from all segments or groups of segments. --Kjoonlee 03:32, 11 February 2007 (UTC)

Is the article wrong? If yes, then it needs to be corrected. Is it correct? In that case, I hope it gets a bit clearer. --Kjoonlee 03:33, 11 February 2007 (UTC)

I'm not sure what you mean by "lexically distinctive", but morphemes can be expressed by suprasegmental features. See, e.g., this on floating tones. Rmalouf (talk) 01:41, 28 February 2014 (UTC)

Oligosynthetic & Isolating Language Summaries

...are not in this article. I would find it most helpful if someone added them. --124.170.35.1 09:24, 10 October 2007 (UTC)

There is some confusion with terminology

In this page, languages are divided in "analytic" and "synthetic" ones, but in the box on the right they're classified as "isolating" and "synthetic".

The correct classification should be:

• Analytic: Isolating (and other kinds?)
• Synthetic: Polysynthetic, Fusional, Agglutinative

Isolating languages are the only subset of analytic languages described on Wikipedia, and the page Analytic language redirects to Isolating language. But "isolating" is a kind of analytic"!

The fact that both the Analytic language page and that the "Analytic" voice in the right box are missing can create some confusion.

I suggest to change "Isolating" with "Analytic" in the right box, and to create a short Analytic language page (in this case, "Isolating" can be indicated as a kind of "Analytic" in the right box).

--Taekwondavide (talk) 22:00, 6 March 2010 (UTC)

New list

I'm trying to put together a tentative list of the languages of the world with the linguistic features of each. It's a huge task and I don't have time to do it on my own. Please come here - User:Bienfuxia/List of languages by linguistic features - and help out a bit, when it's ready I can publish it as a proper article. This is one occasion where wiki is missing something substantive that it should have, please come along and do what you can! Bienfuxia (talk) 07:43, 10 March 2011 (UTC)

Problems with this article

This article relies on the simplistic and now mostly abandoned typological classification of languages into morphological macrotypes such as analytic, isolating, polysynthetic etc. These are not really considered relevant typological categories by any linguists any more. The article should not be structured around these categories. I would recommend describing all of the traditional macrotypes in a single section. Also the only way that these types make sense is by classifying specific words or morphemes as using one or the other type. Noone classifies languages as "being" of X type anymore, since all languages include different types of morphology. WALS is cited as the exponent of this view, but WALS (which is produced by the worlds foremost typologists) is simply describing the actual consensus within linguistics. This is the viewpoint that should be the main viewpoint in the article.User:Maunus ·ʍaunus·snunɐw· 19:44, 28 November 2014 (UTC)

Can you provide some scholars who disagree, and some evidence that they're dominant nowadays? Because despite the fact that I myself find the traditional convention over-simplistic, I've seen very little protestation from scholars. Tezero (talk) 19:52, 28 November 2014 (UTC)

Even Sapir in 1921 called the typology "popular" and then went on to argue that it was little useful. Ive referred to some typology textbooks below, you should look and see how they describe the traditional typology.User:Maunus ·ʍaunus·snunɐw· 20:20, 28 November 2014 (UTC)

No time or spoons to deal with it now, but Fusion of Selected Inflectional Formatives, Exponence of Selected Inflectional Formatives, and the references in both may help with the missing information and maybe (some of) the missing references. The Crab Who Played With The Sea (talk) 20:33, 30 January 2022 (UTC)

Oligosynthesis

Oligosynthesis is not and has never been seriously considered a morphological type. Giving it a section is a lot of undue weight.User:Maunus ·ʍaunus·snunɐw· 19:44, 28 November 2014 (UTC)

Where else should it be listed? It doesn't have to be a large section. Tezero (talk) 19:52, 28 November 2014 (UTC)

It probably shouldnt because it is never mentioned in descriptions of linguistic typology.User:Maunus ·ʍaunus·snunɐw· 20:03, 28 November 2014 (UTC)

Literature

This article should be based on the main linguistic literature about typology. Not willi-nilli picked publications.User:Maunus ·ʍaunus·snunɐw· 19:44, 28 November 2014 (UTC)

It's not "based on" anything. The article had zero citations before I started editing it, so I have no idea where most of it came from. Heck, a lot of what remains on the page is OR that I haven't checked. The citations present are only ones I've sought out to verify claims I foresaw being controversial. Tezero (talk) 19:52, 28 November 2014 (UTC)

I am not blaming you, but making a statement about the current state of the article. I recommend using Comrie's "Language UNiversals and Linguistic Typology", Laurie Bauers "Introducing Linguistic Morpohology", Crofts "Typology and Universals" and Whaley's "Introduction to Typology" to find out how contemporary linguistic typologists work with morphology, and use that knowledge as a basis for structuring the article.User:Maunus ·ʍaunus·snunɐw· 20:03, 28 November 2014 (UTC)

Headmarking/Dependent marking

Johanna Nichols concepts of dependent and headmarking morphology are also morphological types, of a more functional nature, and not based on Sapirs notion of the synthesis/fusion indexes. This distinction is more commonly used in contemporary morphological typology than the traditional macrotypes. It should be included. Whaley's introduction to typology has a good description of both Sapir and Nichols systems in chapter 8.User:Maunus ·ʍaunus·snunɐw· 21:06, 28 November 2014 (UTC)

Polysynthetic subsection: translation requires source

I think that the translation that is marked with falls under you don't need to cite that the sky is blue. Translations are obvious information, no matter how obscure the language (provided it is still alive). Amp2001 (talk) 22:07, 25 June 2021 (UTC)